The objectives of this five-year study are to increase community use of evidence-based prevention programs and subsequently decrease adolescent drug use. The specific aims are to: (1) test the effects of two prevention training interventions with high national diffusion potential (televised training with interactive satellite down-link capability, training + off-site technical assistance, control), on community leader and community adoption, implementation, and dissemination of evidence-based prevention programs; and (2) test the effect of these interventions at the community level, and the effect of a classroom-based media literacy program at the school level, on increasing adolescent attention to the national anti-use campaign, use of local media for promoting prevention messages and programs, and changes in use norms and behaviors. Two research designs are used: at the community level, communities will be randomly assigned from within each of 8 states to one of the three training conditions (5 training segments across two years on needs assessment, community organization, media literacy/advocacy, school, and parent programs; training + assistance, control; N=24 communities); at the school level, two middle/junior high schools within each community will be randomly assigned to a media literacy program on interpreting and counteracting drug use images, use of local media, and social norm expectations; or control; N=48 schools). At the community level, subjects are community leaders (youth/recreation agency, school, parent, policy/government, law enforcement, media, business, special interest; n=25-35 per community; N=840). At the school level, subjects are all 6,7,8th grade students from 2 classes/school and parents of 7th graders; N=5,760). The measurement design is longitudinal, with community leaders and students followed annually for four years with surveys and archival records. Independent variables include intervention and time; covariates include demographic variables, community readiness, and drug use risk; dependent variables include participation in community organization for prevention and use of evidence-based prevention programs and local media for prevention; and drug use norms and drug use. Intervention effects will be analyzed with regression, growth curve, structural equation modeling, time series, and cost-effectiveness analyses, with community, school, and individual as the units of analysis. Results should be generalizable to all communities and cities with satellite down-link capabilities (the majority of U.S. communities). The findings should help determine whether evidence-based programs supported by federal research agencies and Congress can be disseminated to and used by communities at relatively low cost, with a relatively high degree of fidelity, and whether communities can use available mass media constructively to enhance prevention initiatives and anti-use norms.